Most human tumors arise from self-renewing cells of epithelial tissues. Tumors in the lung are the most deadly, with a projected 152,000 deaths in 2003 in the US alone. A strong genetic component contributes to lung cancer risk, as shown by evidence of familial clustering, and the fact that only 10-20% of smokers develop lung cancer. Identification of genetic variants that confer susceptibility to lung cancer will enable prediction of individual risk, and facilitate the development of novel therapeutic or preventive agents. A particular species of mouse (Musspretus) is genetically resistant to development of lung cancer, due to the presence of germline polymorphisms that confer resistance. Crosses between sensitive and resistant mouse strains will be used in a genetic approach to identify lung tumor resistance genes. This analysis will be facilitated by the availability of an extensive database and tissue/tumor bank derived from a large backcross between the lung tumor-susceptible KrasM2 mice on the FVB background, with the resistant Mus spretus species. Data are already available both on the genomic changes associated with lung cancer initiation and progression, and on the genetic basis of lung cancer susceptibility. These data have implicated the Kras2 gene as a major determinant of lung tumor development and susceptibility. This particular ras gene family member is frequently mutated in both human and mouse lung tumors, and is one of the candidates for the major mouse lung tumor susceptibility locus Past. Two engineered mouse models will be used to test the role of Kras2 in lung tumor development and susceptibility. One involves a knock-in of a mutant Kras2 allele into the endogenous locus (/<rasM2 mice), resulting in 100% incidence of spontaneous lung tumors without carcinogen treatment. An additional model (HrasKI mice) involves a knock-in of the Hras1coding sequence into the Kras2 locus. Crosses between these strains and wild type mice of different genetic backgrounds will be used to analyze the requirement for Kras2 mutations in lung cancer, and to define the role of Kras2 in lung tumor susceptibility. Linkage analysis of lung tumor susceptibility has revealed the presence of several additional lung tumor resistance or susceptibility genes, including a locus on mouse chromosome 7 (Ltr1) that suppresses lung tumor formation in congenic mice. A combination of linkage data, genomic alterations in lung tumors, and gene expression profiles will be used to refine the regions containing the lung tumor resistance genes Ltr1 and the Pulmonary Adenoma Resistance gene Par1, which is known to interact with Pasf.Candidate genes will be selected by analysis of allele-specific genetic alterations in tumors using genome wide high-density BAG arrays, together with gene expression microarrays to profile both normal tissues and tumors from backcross animals. The availability of these novel models and reagents will help us to address several long standing questions related to the genetic basis of lung tumor susceptibility.